Syrellaris wrote...
TheRealMithril wrote...
Relwyn wrote...
As I work in a scientific field for a living, a sample of 50 000 is overkill for any scientific study when extrapolating the results of the statistics onto a larger population. Normally a study has a few hundred or a few thousand individuals participating because it is enough to make conclusions based on the statistical analysis.
A sample size of tens of thousands is simply put overkill in terms of statistics. Studies with such a sample size are on a continental or national scale. While the poll itself is simply put a poll and nothing more, it's limited in some ways - however: Anyone who sees it at Bioware is probably lighting a cigarette with shaky hands because they know that the poll is a good indication of the public's view of the game's ending.
To do as IGN and other organizations (including Bioware) have done and call it a minority or try to downplay the size of the community which thinks the ending is terrible is to be intellectually disingenuous/dishonest.![]()
While they do their PR-shell game for now (which they can only keep up for so long, sooner or later they'll have to come clean), I'll shamelessely Hold the Line.
Kudos, this very point was raised by many of us when attempts were made to minimalize the effects of the polls
I'll repeat what I also said (in a similar situation, because this keep getting lost in the steady stream of other posts) in regards to this point: The polls (even if badly made) are so large that even with a heathly 'rebate' there would still be more than 75% who are against the endings. The polls are so huge that they would tell you who the next president is, it is that accurate.
I don't mean to steal your point (just to make it clear) only to add support to it
Again, kudos.
They would have probable cause if they were accurate. Which the poll on BSN is not. Even if its 50 000 votes, the problem with the poll is people can constantly revote and each vote would count. Any scientist, evne the one you quoted or any other person that has to deal with polls and statistics, would dismiss a Inaccurate poll right away.
Not a pure scientist but in the field, so here's my interpritation
1) Polls like this will be biased towards the strong opinions, which most people who played the game and liked the ending will not have. Therefore even though 90% (don't know the exact number) disliked the ending to extrapolate that out over the millions is not accurate. However, while this is the case it is totally inaccurate to say that 'only people who voted disliked', that is totally wrong, personally I would say, if the poll indicates 90%, then somwhere between 50-70% most likely dislike or strongly dislike the ending, that however is purly a hypothisis with no dirrect scientific proof.
2) The poll is probably bias, its extreemly hard to word questions which don't lead you to a given answer, its actually a bit of an art getting the questions right. That said, everyone knows the answer they would give no matter the question so i feel in this case that is no longer a vaild point.
3)I also think that the PR people forgot we aren't all 12 years old and easily swayed by faulty logic (I mean no offence to 12 year olds, just thats what non-gamers often think of us)
4) 50,000 is massivly more than needed to show scientific treanding. I have heard of experiments with samples in the 10's being given academic/media credence.




Ce sujet est fermé
Retour en haut




